Image reading devices for use in banking terminal devices, such as check readers, bill counters, bill sorters, and automated teller machines (ATMs), read, as electrical information, security information printed or imprinted on media such as bills, checks, and securities, and, based on the read information, determine a monetary amount or perform authentication.
One method for converting information of such a medium, which is an object to be read, into electrical information is to convert a two-dimensional image into an electrical signal, such as in a digital camera. Another method is to use a linear sensor. The linear sensor is an array of photoelectric conversion elements arranged to have the same width as the medium. The linear sensor obtains a one-dimensional image by illuminating the medium and passing the light reflected by the illuminated medium through a lens array to form an image, and then obtains a two-dimensional image by movement of the medium in a direction perpendicular to the array of photoelectric conversion elements. The method of using the linear sensor leads to size reduction of the overall device and is thus commonly adopted.
When the object to be read includes different information on the front and back, both the front and back image information are needed to improve the accuracy in reading of the object to be read. However, double-sided reading with a single sensor by reversing the paper money takes additional time for conveyance, and also takes twice the processing time for a single object to be read, compared with the processing time taken for single-sided reading. Moreover, arrangement of more space is required to provide a conveyance path for reversing the front-back of the object to be read, and such arrangement is not suitable for reduction in size. Thus, another arrangement is adopted for double-sided reading in one conveyance, in which a sensor for reading the front and a sensor for reading the back are provided. For more sophisticated identification, print misregistration between the front and back and positioning of a watermark image using a transmission light source can be adjusted by synthesis of front and back image information and processing of the image.
Patent Literature 1 discloses a correction method for an image reading device that calculates an amount of reading distortion for each read position within a read area by comparison between correction patterns and that corrects the reading distortion of image data based on the calculated amount of reading distortion.
Patent Literature 2 discloses an image formation device for alignment of front and back images by eliminating not only misalignment due to shift of front and back images but also deviations of magnifications of the images due to expansion and contraction of recording paper.
Patent Literature 3 discloses a correction method for measurement of a distance between any two positions on an image of a document based on image data obtained by reading the image of the document. This method for an image measurement device involves calculating a value of difference between actual measurement values of positions of specific points on a specific pattern of a correction chart and positions of the specific points from image data obtained by reading the correction chart. Then the measurement of the distance between any two positions on the image of the document is corrected based on the data of the value of difference.